1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
|
#define LIMIT -999
//#define TRACE
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <math.h>
#include <sys/time.h>
#include <omp.h>
#define OPENMP
//#define NUM_THREAD 4
#define BLOCK_SIZE 16
////////////////////////////////////////////////////////////////////////////////
// declaration, forward
void runTest( int argc, char** argv);
// Returns the current system time in microseconds
long long get_time()
{
struct timeval tv;
gettimeofday(&tv, NULL);
return (tv.tv_sec * 1000000) + tv.tv_usec;
}
#ifdef OMP_OFFLOAD
#pragma omp declare target
#endif
int maximum( int a,
int b,
int c){
int k;
if( a <= b )
k = b;
else
k = a;
if( k <=c )
return(c);
else
return(k);
}
#ifdef OMP_OFFLOAD
#pragma omp end declare target
#endif
int blosum62[24][24] = {
{ 4, -1, -2, -2, 0, -1, -1, 0, -2, -1, -1, -1, -1, -2, -1, 1, 0, -3, -2, 0, -2, -1, 0, -4},
{-1, 5, 0, -2, -3, 1, 0, -2, 0, -3, -2, 2, -1, -3, -2, -1, -1, -3, -2, -3, -1, 0, -1, -4},
{-2, 0, 6, 1, -3, 0, 0, 0, 1, -3, -3, 0, -2, -3, -2, 1, 0, -4, -2, -3, 3, 0, -1, -4},
{-2, -2, 1, 6, -3, 0, 2, -1, -1, -3, -4, -1, -3, -3, -1, 0, -1, -4, -3, -3, 4, 1, -1, -4},
{ 0, -3, -3, -3, 9, -3, -4, -3, -3, -1, -1, -3, -1, -2, -3, -1, -1, -2, -2, -1, -3, -3, -2, -4},
{-1, 1, 0, 0, -3, 5, 2, -2, 0, -3, -2, 1, 0, -3, -1, 0, -1, -2, -1, -2, 0, 3, -1, -4},
{-1, 0, 0, 2, -4, 2, 5, -2, 0, -3, -3, 1, -2, -3, -1, 0, -1, -3, -2, -2, 1, 4, -1, -4},
{ 0, -2, 0, -1, -3, -2, -2, 6, -2, -4, -4, -2, -3, -3, -2, 0, -2, -2, -3, -3, -1, -2, -1, -4},
{-2, 0, 1, -1, -3, 0, 0, -2, 8, -3, -3, -1, -2, -1, -2, -1, -2, -2, 2, -3, 0, 0, -1, -4},
{-1, -3, -3, -3, -1, -3, -3, -4, -3, 4, 2, -3, 1, 0, -3, -2, -1, -3, -1, 3, -3, -3, -1, -4},
{-1, -2, -3, -4, -1, -2, -3, -4, -3, 2, 4, -2, 2, 0, -3, -2, -1, -2, -1, 1, -4, -3, -1, -4},
{-1, 2, 0, -1, -3, 1, 1, -2, -1, -3, -2, 5, -1, -3, -1, 0, -1, -3, -2, -2, 0, 1, -1, -4},
{-1, -1, -2, -3, -1, 0, -2, -3, -2, 1, 2, -1, 5, 0, -2, -1, -1, -1, -1, 1, -3, -1, -1, -4},
{-2, -3, -3, -3, -2, -3, -3, -3, -1, 0, 0, -3, 0, 6, -4, -2, -2, 1, 3, -1, -3, -3, -1, -4},
{-1, -2, -2, -1, -3, -1, -1, -2, -2, -3, -3, -1, -2, -4, 7, -1, -1, -4, -3, -2, -2, -1, -2, -4},
{ 1, -1, 1, 0, -1, 0, 0, 0, -1, -2, -2, 0, -1, -2, -1, 4, 1, -3, -2, -2, 0, 0, 0, -4},
{ 0, -1, 0, -1, -1, -1, -1, -2, -2, -1, -1, -1, -1, -2, -1, 1, 5, -2, -2, 0, -1, -1, 0, -4},
{-3, -3, -4, -4, -2, -2, -3, -2, -2, -3, -2, -3, -1, 1, -4, -3, -2, 11, 2, -3, -4, -3, -2, -4},
{-2, -2, -2, -3, -2, -1, -2, -3, 2, -1, -1, -2, -1, 3, -3, -2, -2, 2, 7, -1, -3, -2, -1, -4},
{ 0, -3, -3, -3, -1, -2, -2, -3, -3, 3, 1, -2, 1, -1, -2, -2, 0, -3, -1, 4, -3, -2, -1, -4},
{-2, -1, 3, 4, -3, 0, 1, -1, 0, -3, -4, 0, -3, -3, -2, 0, -1, -4, -3, -3, 4, 1, -1, -4},
{-1, 0, 0, 1, -3, 3, 4, -2, 0, -3, -3, 1, -1, -3, -1, 0, -1, -3, -2, -2, 1, 4, -1, -4},
{ 0, -1, -1, -1, -2, -1, -1, -1, -1, -1, -1, -1, -1, -1, -2, 0, 0, -2, -1, -1, -1, -1, -1, -4},
{-4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, -4, 1}
};
double gettime() {
struct timeval t;
gettimeofday(&t,NULL);
return t.tv_sec+t.tv_usec*1e-6;
}
////////////////////////////////////////////////////////////////////////////////
// Program main
////////////////////////////////////////////////////////////////////////////////
int
main( int argc, char** argv)
{
runTest( argc, argv);
return EXIT_SUCCESS;
}
void usage(int argc, char **argv)
{
fprintf(stderr, "Usage: %s <max_rows/max_cols> <penalty> <num_threads>\n", argv[0]);
fprintf(stderr, "\t<dimension> - x and y dimensions\n");
fprintf(stderr, "\t<penalty> - penalty(positive integer)\n");
fprintf(stderr, "\t<num_threads> - no. of threads\n");
exit(1);
}
void nw_optimized(int *input_itemsets, int *output_itemsets, int *referrence,
int max_rows, int max_cols, int penalty)
{
#ifdef OMP_OFFLOAD
int transfer_size = max_rows * max_cols;
#pragma omp target data map(to: max_cols, penalty, referrence[0:transfer_size]) map(input_itemsets[0:transfer_size])
{
#pragma omp target
#endif
for( int blk = 1; blk <= (max_cols-1)/BLOCK_SIZE; blk++ )
{
#ifdef OPENMP
#pragma omp parallel for schedule(static) shared(input_itemsets, referrence) firstprivate(blk, max_rows, max_cols, penalty)
#endif
for( int b_index_x = 0; b_index_x < blk; ++b_index_x)
{
int b_index_y = blk - 1 - b_index_x;
int input_itemsets_l[(BLOCK_SIZE + 1) *(BLOCK_SIZE+1)] __attribute__ ((aligned (64)));
int reference_l[BLOCK_SIZE * BLOCK_SIZE] __attribute__ ((aligned (64)));
// Copy referrence to local memory
for ( int i = 0; i < BLOCK_SIZE; ++i )
{
#pragma omp simd
for ( int j = 0; j < BLOCK_SIZE; ++j)
{
reference_l[i*BLOCK_SIZE + j] = referrence[max_cols*(b_index_y*BLOCK_SIZE + i + 1) + b_index_x*BLOCK_SIZE + j + 1];
}
}
// Copy input_itemsets to local memory
for ( int i = 0; i < BLOCK_SIZE + 1; ++i )
{
#pragma omp simd
for ( int j = 0; j < BLOCK_SIZE + 1; ++j)
{
input_itemsets_l[i*(BLOCK_SIZE + 1) + j] = input_itemsets[max_cols*(b_index_y*BLOCK_SIZE + i) + b_index_x*BLOCK_SIZE + j];
}
}
// Compute
for ( int i = 1; i < BLOCK_SIZE + 1; ++i )
{
for ( int j = 1; j < BLOCK_SIZE + 1; ++j)
{
input_itemsets_l[i*(BLOCK_SIZE + 1) + j] = maximum( input_itemsets_l[(i - 1)*(BLOCK_SIZE + 1) + j - 1] + reference_l[(i - 1)*BLOCK_SIZE + j - 1],
input_itemsets_l[i*(BLOCK_SIZE + 1) + j - 1] - penalty,
input_itemsets_l[(i - 1)*(BLOCK_SIZE + 1) + j] - penalty);
}
}
// Copy results to global memory
for ( int i = 0; i < BLOCK_SIZE; ++i )
{
#pragma omp simd
for ( int j = 0; j < BLOCK_SIZE; ++j)
{
input_itemsets[max_cols*(b_index_y*BLOCK_SIZE + i + 1) + b_index_x*BLOCK_SIZE + j + 1] = input_itemsets_l[(i + 1)*(BLOCK_SIZE+1) + j + 1];
}
}
}
}
printf("Processing bottom-right matrix\n");
#ifdef OMP_OFFLOAD
#pragma omp target
#endif
for ( int blk = 2; blk <= (max_cols-1)/BLOCK_SIZE; blk++ )
{
#ifdef OPENMP
#pragma omp parallel for schedule(static) shared(input_itemsets, referrence) firstprivate(blk, max_rows, max_cols, penalty)
#endif
for( int b_index_x = blk - 1; b_index_x < (max_cols-1)/BLOCK_SIZE; ++b_index_x)
{
int b_index_y = (max_cols-1)/BLOCK_SIZE + blk - 2 - b_index_x;
int input_itemsets_l[(BLOCK_SIZE + 1) *(BLOCK_SIZE+1)] __attribute__ ((aligned (64)));
int reference_l[BLOCK_SIZE * BLOCK_SIZE] __attribute__ ((aligned (64)));
// Copy referrence to local memory
for ( int i = 0; i < BLOCK_SIZE; ++i )
{
#pragma omp simd
for ( int j = 0; j < BLOCK_SIZE; ++j)
{
reference_l[i*BLOCK_SIZE + j] = referrence[max_cols*(b_index_y*BLOCK_SIZE + i + 1) + b_index_x*BLOCK_SIZE + j + 1];
}
}
// Copy input_itemsets to local memory
for ( int i = 0; i < BLOCK_SIZE + 1; ++i )
{
#pragma omp simd
for ( int j = 0; j < BLOCK_SIZE + 1; ++j)
{
input_itemsets_l[i*(BLOCK_SIZE + 1) + j] = input_itemsets[max_cols*(b_index_y*BLOCK_SIZE + i) + b_index_x*BLOCK_SIZE + j];
}
}
// Compute
for ( int i = 1; i < BLOCK_SIZE + 1; ++i )
{
for ( int j = 1; j < BLOCK_SIZE + 1; ++j)
{
input_itemsets_l[i*(BLOCK_SIZE + 1) + j] = maximum( input_itemsets_l[(i - 1)*(BLOCK_SIZE + 1) + j - 1] + reference_l[(i - 1)*BLOCK_SIZE + j - 1],
input_itemsets_l[i*(BLOCK_SIZE + 1) + j - 1] - penalty,
input_itemsets_l[(i - 1)*(BLOCK_SIZE + 1) + j] - penalty);
}
}
// Copy results to global memory
for ( int i = 0; i < BLOCK_SIZE; ++i )
{
#pragma omp simd
for ( int j = 0; j < BLOCK_SIZE; ++j)
{
input_itemsets[max_cols*(b_index_y*BLOCK_SIZE + i + 1) + b_index_x*BLOCK_SIZE + j + 1] = input_itemsets_l[(i + 1)*(BLOCK_SIZE+1) + j +1];
}
}
}
}
#ifdef OMP_OFFLOAD
}
#endif
}
////////////////////////////////////////////////////////////////////////////////
//! Run a simple test for CUDA
////////////////////////////////////////////////////////////////////////////////
void
runTest( int argc, char** argv)
{
int max_rows, max_cols, penalty;
int *input_itemsets, *output_itemsets, *referrence;
//int *matrix_cuda, *matrix_cuda_out, *referrence_cuda;
//int size;
int omp_num_threads;
// the lengths of the two sequences should be able to divided by 16.
// And at current stage max_rows needs to equal max_cols
if (argc == 4)
{
max_rows = atoi(argv[1]);
max_cols = atoi(argv[1]);
penalty = atoi(argv[2]);
omp_num_threads = atoi(argv[3]);
}
else{
usage(argc, argv);
}
max_rows = max_rows + 1;
max_cols = max_cols + 1;
referrence = (int *)malloc( max_rows * max_cols * sizeof(int) );
input_itemsets = (int *)malloc( max_rows * max_cols * sizeof(int) );
output_itemsets = (int *)malloc( max_rows * max_cols * sizeof(int) );
if (!input_itemsets)
fprintf(stderr, "error: can not allocate memory");
srand ( 7 );
for (int i = 0 ; i < max_cols; i++){
for (int j = 0 ; j < max_rows; j++){
input_itemsets[i*max_cols+j] = 0;
}
}
printf("Start Needleman-Wunsch\n");
for( int i=1; i< max_rows ; i++){ //please define your own sequence.
input_itemsets[i*max_cols] = rand() % 10 + 1;
}
for( int j=1; j< max_cols ; j++){ //please define your own sequence.
input_itemsets[j] = rand() % 10 + 1;
}
for (int i = 1 ; i < max_cols; i++){
for (int j = 1 ; j < max_rows; j++){
referrence[i*max_cols+j] = blosum62[input_itemsets[i*max_cols]][input_itemsets[j]];
}
}
for( int i = 1; i< max_rows ; i++)
input_itemsets[i*max_cols] = -i * penalty;
for( int j = 1; j< max_cols ; j++)
input_itemsets[j] = -j * penalty;
//Compute top-left matrix
printf("Num of threads: %d\n", omp_num_threads);
printf("Processing top-left matrix\n");
long long start_time = get_time();
nw_optimized( input_itemsets, output_itemsets, referrence,
max_rows, max_cols, penalty );
long long end_time = get_time();
printf("Total time: %.3f seconds\n", ((float) (end_time - start_time)) / (1000*1000));
#define TRACEBACK
#ifdef TRACEBACK
FILE *fpo = fopen("result.txt","w");
fprintf(fpo, "print traceback value GPU:\n");
for (int i = max_rows - 2, j = max_rows - 2; i>=0, j>=0;){
int nw, n, w, traceback;
if ( i == max_rows - 2 && j == max_rows - 2 )
fprintf(fpo, "%d ", input_itemsets[ i * max_cols + j]); //print the first element
if ( i == 0 && j == 0 )
break;
if ( i > 0 && j > 0 ){
nw = input_itemsets[(i - 1) * max_cols + j - 1];
w = input_itemsets[ i * max_cols + j - 1 ];
n = input_itemsets[(i - 1) * max_cols + j];
}
else if ( i == 0 ){
nw = n = LIMIT;
w = input_itemsets[ i * max_cols + j - 1 ];
}
else if ( j == 0 ){
nw = w = LIMIT;
n = input_itemsets[(i - 1) * max_cols + j];
}
else{
}
//traceback = maximum(nw, w, n);
int new_nw, new_w, new_n;
new_nw = nw + referrence[i * max_cols + j];
new_w = w - penalty;
new_n = n - penalty;
traceback = maximum(new_nw, new_w, new_n);
if(traceback == new_nw)
traceback = nw;
if(traceback == new_w)
traceback = w;
if(traceback == new_n)
traceback = n;
fprintf(fpo, "%d ", traceback);
if(traceback == nw )
{i--; j--; continue;}
else if(traceback == w )
{j--; continue;}
else if(traceback == n )
{i--; continue;}
else
;
}
fclose(fpo);
#endif
free(referrence);
free(input_itemsets);
free(output_itemsets);
}
|
